Gareth

Key Skills

Software

Don't disclose

Top Subject Matter

Hospitality Domain Expertise

Operations Domain Expertise

Sales Domain Expertise

Top Data Types

Text

Document

Top Task Types

RLHF

Entity (NER) Classification

Freelancer Overview

AI Integration Builder & Prompt Developer (O= Independent Venture). Brings 19+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Internal, Proprietary Tooling, and Don't disclose. Education includes Beach Lifeguard Training and Assessment, RNLI (2019) and Royal Navy Communications Specialist Certification, Royal Navy (2012). AI-training focus includes data types such as Text and labeling workflows including RLHF, Named Entity Recognition, and Evaluation.

Labeling Experience

AI Integration Builder & Prompt Developer (O= Independent Venture)

TextRLHFEntity (NER) Classification

Evaluated and refined LLM outputs by comparing prompt variants, identifying edge cases, and adjusting instructions for consistent behavior. Produced multilingual AI content and assessed tone, accuracy, and cultural fit across English, German, and Portuguese. Documented model failure modes and prompt changes to support reproducibility and iterative quality improvement. • Data work included preference-ranking style comparisons and RLHF-adjacent feedback workflows • LLM evaluation focused on factual accuracy, register, and cultural appropriateness • Iteration included prompt testing, output comparison, and failure-mode mitigation • Generated structured responses for a guest-assistant workflow using multi-step prompting

2024 - Present

Corporate Trainer — DACH Region (Various Clients)

Don't discloseText

Designed and delivered training programs and evaluated delegate outputs against defined rubrics. Provided structured, calibrated feedback aimed at improving correctness and communication quality. Supported bilingual delivery in English and German while handling cultural and register nuances. • Evaluation work functioned as a quality-rating loop for trainee responses • Feedback was structured using rubrics and performance criteria • Cross-cultural instruction improved language appropriateness in outputs • Training emphasis mirrored AI response evaluation and refinement practices

2019 - 2023

Senior Beach Lifeguard & Trainer — RNLI

Don't discloseText

Trained junior lifeguards using unambiguous instructional content and criteria-based assessment. Evaluated trainee performance against defined requirements and provided structured guidance to improve execution quality. Applied emphasis on clarity and accuracy under operational constraints. • Assessment used defined criteria to judge correctness of actions/responses • Instructional clarity and ambiguity handling supported reliable output evaluation • Structured feedback practices parallel reviewing and refining responses for quality

2012 - 2019

Education

Work History